Beta_diversity_MILK
Chiara Gini
27/12/2021
Beta diversity
From 16S rRNA-gene sequencing of cow milk microbiome samples: - 4 timepoints for milk samples and two treatments (oil, treated/untreated)
Between-sample variability from 16S rRna-gene sequencing data.
Clustering
PCA
Clustering by timepoint (PCA):
Clustering by treatment (PCA): after removing samples from timepoint 0 # BECAUSE WE KNOW THAT AT TIMEPOINT 0 WE SHOULD NOT HAVE DIFFERENCES BETWEEN GROUPS
Clustering by treatment (PCA): by timepoints
Time-point 28
## Warning in data.frame(..., check.names = FALSE): row names were found from a
## short variable and have been discarded
Time-point 21
## Warning in data.frame(..., check.names = FALSE): row names were found from a
## short variable and have been discarded
Time-point 8
## Warning in data.frame(..., check.names = FALSE): row names were found from a
## short variable and have been discarded
Time-point 0
## Warning in data.frame(..., check.names = FALSE): row names were found from a
## short variable and have been discarded
MDS
Clustering by treatment (MDS): NMDS1 vs NMDS2
Clustering by timepoint (MDS):
NMDS1 vs NMDS2
Clustering by type (MDS): NMDS1 vs NMDS3
Clustering by type (MDS): NMDS2 vs NMDS3
Clustering by treatment by timepoint (MDS):
Significance of between-group distances
Significance values based on permuted analysis of variance (999 permutations), repeated 100 times.
- between treatments: p-value = 0.05639 (std = 0.0068533)
- between types: p-value = 0.0069537 (std = 0.0083428)
- between treatments and type = p-value = 0.04468 (std = 0.0064321)
Below the ANOVA table from the model:
\[ X = \mu + treatment + timepoint + treatment*type + e \]
| Df | SumsOfSqs | MeanSqs | F.Model | R2 | Pr(>F) | |
|---|---|---|---|---|---|---|
| treatment | 1 | 0.3754833 | 0.3754833 | 1.702507 | 0.0256107 | 0.041 |
| timepoint | 3 | 1.2733871 | 0.4244624 | 1.924586 | 0.0868544 | 0.004 |
| residuals | 59 | 13.0122919 | 0.2205473 | NA | 0.8875348 | NA |
| total | 63 | 14.6611623 | NA | NA | 1.0000000 | NA |
going 3D
## Run 0 stress 0.07511949
## Run 1 stress 0.07982514
## Run 2 stress 0.07979209
## Run 3 stress 0.07508381
## ... New best solution
## ... Procrustes: rmse 0.006178098 max resid 0.03090521
## Run 4 stress 0.07511025
## ... Procrustes: rmse 0.005230696 max resid 0.03103953
## Run 5 stress 0.08117002
## Run 6 stress 0.07511956
## ... Procrustes: rmse 0.006163733 max resid 0.03118073
## Run 7 stress 0.08845616
## Run 8 stress 0.07508376
## ... New best solution
## ... Procrustes: rmse 0.0001836155 max resid 0.0009432191
## ... Similar to previous best
## Run 9 stress 0.09210695
## Run 10 stress 0.07541573
## ... Procrustes: rmse 0.01042587 max resid 0.06485814
## Run 11 stress 0.08579224
## Run 12 stress 0.07510211
## ... Procrustes: rmse 0.003036263 max resid 0.0197986
## Run 13 stress 0.07979847
## Run 14 stress 0.07511062
## ... Procrustes: rmse 0.005221525 max resid 0.03095948
## Run 15 stress 0.08233311
## Run 16 stress 0.08879875
## Run 17 stress 0.08829215
## Run 18 stress 0.07511165
## ... Procrustes: rmse 0.005246247 max resid 0.03101884
## Run 19 stress 0.07508407
## ... Procrustes: rmse 0.0001698039 max resid 0.00101293
## ... Similar to previous best
## Run 20 stress 0.07512631
## ... Procrustes: rmse 0.005689686 max resid 0.03113833
## *** Solution reached